Although atrial fibrillation (AF) is the most prevalent cardiac arrhythmia requiring antiarrhythmic drug therapy, response in an individual patient is highly variable with frequent AF recurrence. The limited success of drug therapy has been attributed in part to failure to target therapy to underlying mechanisms. Recent experimental work from our group and others in mouse models suggests that Ca leak via type 2 ryanodine receptor (RyR2) Ca release channels is one important mechanism responsible for triggering paroxysmal AF. While Ca leak has been documented in atrial tissue from AF patients, the causal relationship to human AF remains controversial. Hence, proposed here will test our overarching hypothesis that leaky RyR2 channels confer AF risk that can be targeted therapeutically in humans. Our published and preliminary data demonstrate that the R-enantiomer of propafenone is the most potent inhibitor of RyR2 Ca release among clinically approved antiarrhythmic drugs. R-propafenone was strikingly effective in suppressing Ca-triggered focal AF in calsequestrin null mice (Casq2-/-), whereas S-propafenone that largely lacks RyR2 blocking properties was significantly less effective. Since R- and S-propafenone are equipotent Na channel blockers and racemic propafenone is used clinically, the two propafenone enantiomers can be used as tools to determine the contribution of leaky RyR2 channels to AF risk in humans, enabling us to translate findings from mice to patients. In humans, common AF-associated 4q25 single nucleotide polymorphisms (SNPs) near the paired-like homeodomain transcription factor 2 (PITX2) are currently the strongest genetic markers of AF risk. While the functional effects of the SNPs remain controversial, our published data show that individuals carrying 4q25 risk alleles respond better to Class IC drugs flecainide and propafenone, both of which also inhibit RyR2 channels. In contrast, Class III drugs that lack RyR2 blocking properties were less effective in 4q25 carriers. These results raise the exciting possibility to be tested below that 4q25 risk alleles are markers for leaky RyR2 channels in humans with AF. Hence, Aim 1 will determine whether Ca leak contributes to the underlying pro-arrhythmic mechanism in diverse mouse models with inducible AF: Pitx2 haploinsufficient mice (Pitx2+/-), mice carrying human AF mutations in the cardiac Na channel (Scn5a-D1275N) in the atrial natriuretic peptide (mut-NPPA), and in troponin T that causes atrial hypertrophy and fibrosis (TnT-F110I). Aim 2 will test the hypothesis that the underlying electrophysiological mechanisms predict response to drug therapy in mice. The clinical trial in Aim 3 will test whether results from murine AF models predict antiarrhythmic drug efficacy in humans with paroxysmal AF, and whether the 4q25 risk alleles can identify AF patients who will benefit from RyR2 channel block. Accomplishing the aims could provide proof of concept that mouse AF research is translatable to humans, and that drug testing at the time of AF ablation in the clinical EP lab may help individualize AF drug therapy.